Problematic alcohol use emerges early and poses tremendous burden. There is an urgent need to understand who is vulnerable, and why, to facilitate accurate detection, prevention, and treatment. To address this critical issue, a more precise mechanistic understanding of behavioral-brain risk is needed. Data from my graduate research suggests that one source of risk for problematic alcohol use may be heightened reactivity to uncertain threat (U-threat), meaning threat that is ambiguous, temporally unpredictable or uncertain in its intensity. To date, however, no study has directly tested whether reactivity to U-threat functions as a risk factor. The neural processes that mediate this potential source of risk are also unknown; although, there is evidence to suggest that a coordinated frontolimbic circuit, referred to as the Anticipatory Anxiety Network (AAN), modulates U-threat responding. Given that establishing a validated biobehavioral risk model, and elucidating mechanisms underlying excessive alcohol use, has the potential to significantly improve clinical efforts, the proposed study combines behavioral and neural measures, with a prospective design, to test whether reactivity to U-threat is a phenotypic risk factor for problematic alcohol use. This study will enroll 150 healthy, emerging adults, ages 17-19, prior to the onset of alcohol use problems and assess their behavioral (i.e., startle eyeblink) and neural (i.e., functional magnetic resonance imaging [fMRI]) reactivity to U-threat as indices of potential risk. All individuals will subsequently be tracked every 3 mos. for 1-2 years to ascertain onset of problematic alcohol use and escalation in drinking behaviors. Data from this project will allow for an innovative and well-controlled test of whether behavioral reactivity to U-threat is associated with AAN dysfunction (Aim 1), and whether it is a risk factor for problematic alcohol use that is mediated by AAN functioning (Aim 2). In addition, the current study will test the discriminative validity of reactivity to U-threat as a risk factor by also examining whether behavioral and neural reactivity to uncertain rewards (U-reward), which similarly elicit anticipation, is also a risk phenotype for problematic alcohol use (Aim 3). This study concurrently provides an excellent opportunity to receive mentor-directed, hands-on training in several key areas necessary to refine my knowledge and skill-sets. Specifically, I propose a training plan that will focus on three new domains: 1) advanced cognitive behavioral neuroscience and task development; 2) behavioral and subjective laboratory assessments in the context of a longitudinal study design with young adults; and 3) biostatistics. This study coupled with completion of the training goals will effectively propel me towards my long-term goal of an independent career in the translational neuroscience of alcoholism, and will lay the foundation for several future, high- impact R01 studies focused on the prevention and intervention of alcohol use disorder.